Khan Academy Pedagogical Perspective More Important Than User Needs?

I read some of your posts in the other topic and agree with you on many points. I am 67 years old (closer to my 68th birthday than my 67th) and retired from a long career in IT (mostly COBOL programming on mainframes) for almost 3 years. I have been doing Khan Academy since my retirement. I have done every math course from Kindergarten through Multivariable Calculus except for Differential equations and Math for fun and glory (which I am intentionally saving for last). However, I need to go back and do some of the content that was added after I did the courses. I'm doing Science now. I've completed all the Physics and most of the Cosmology and astronomy material and am now doing Chemistry. I am frustrated by the lack of a reset button, too, but that's off-topic for this topic.

I believe I was a victim of certain pedagogy methods in elementary school. I don't have time to post them now, but I do want to do this eventually.

One bit of advice - If you are advising people on language to use to avoid sounding like a 5th grader, please use "going to" instead of "gonna", "have got to" or "must" instead of "gotta", "want to" instead of "wanna", and "yes" instead of "yeah" in your postings. If I have you confused with someone else who posted recently in the other topic, I apologize. Also, the person who posted about this over there used either "wanna", "gonna", or "gotta", but I don't remember which of the three it was.

I use "yeah" sometimes, but I don't use the rest.  I try to write very professionally, because I do write professionally enough for it to matter.  (If you do notice me using any of those, feel free to call me out.  I definitely don't want poor writing to get in the way of others understanding what I am trying to express!)

Thank you for your input.  I was in public school until halfway through 6th grade.  I was both failing and in the Talented and Gifted program (and doing very well there), in large part because public schools don't know what to do with students that get their work done faster than everyone else.  I just quit doing my work when I realized that I was going to be punished with more work if I got it done early.  (No sense taking the risk.)  My mom decided to home school me, and the lack of inefficient, and to be honest cruel, pedagogy allowed me and motivated me to accel.  I fared much better in college, where learning is more self driven.

I've been thinking the same thing.  I think Sal himself would understand and agree with my concerns, but clearly he isn't running the show anymore and has given control over to people that I think are just people trained in American teaching methods who have no clue how people actually learn.  I don't even know how I would contact him though.  I am hoping that somehow this and/or the articles I have written on my blog make it back to him, but I am not holding my breath.

I was an intelligent child in elementary school, but I had some teachers that really did not like smart kids (especially girls). In elementary school back then, we had the same teacher for all academic subjects, and many of them took an integrated approach to teaching their subjects.

For example, when we did word problems in math, to show that we were also mastering language arts, we had to label the results in our answers. If it was a problem that involved money, we had to show the answer as dollars and cents. If it involved anything else, we had to write the label next to the answer. For example: John had two apples. He ate one. How many apples did John have left? It was an incorrect answer to just write "1". It had to be "1 apple". I got marked wrong on the following problem I still remember to this day: Mary has 25 cents. It costs 5 cents to ride the merry-go-round. How many times can Mary ride the merry-go-round? I wrote "5 rides". It was marked wrong because it was supposed to be "5 times". That was in third grade.

I was a pretty good speller. In that same third grade year, I only got one answer marked wrong on a spelling test that whole school year. Here's why. For our elementary school spelling tests, not just that year but the others, even through something called junior high that is usually called middle school these days, the teacher would say the word, use it in a sentence, and then say the word again. We were supposed to wait until she or he finished this whole process before writing the word, but most of us didn't need that sentence. One week in third grade, one of our words was "cent", as in the value of one penny. We had learned about homonyms by then and knew that there were two other words that sound like "cent" - "sent" and "scent". That Friday, when we had our spelling test, when she got to "cent", our teacher said, "Sent. Mother sent me to the store. Sent." Most of us raised our heads and hands and said a collective, "Huh? You have the wrong word!" By then, I had already written "cent", in pencil, because we didn't get to use ink until 4th or 5th grade. No problem, I could just erase it and write "sent". But there was a problem. The teacher told us that she did this deliberately, to see who was following directions and not writing the word until she finished her delivery. She also said that any answer of "cent" would be marked wrong as would - get this - any answer in which it was clear that "cent" had been written, erased, and replaced with "sent". Third graders do not erase very neatly, and I was no exception. Of course, I got it wrong, and a strong temper tantrum on my part did not get her to change her mind. It ruined a perfect record for me.

The following year, my fourth grade teacher also tried a dirty trick to trip up the good spellers in our class. From 2nd-4th grade in my school, we used grade-specific spelling books in the same series in which the same process was followed each week. Monday, we read a story that introduces each word in context. Tuesday, we did some fill-in-the-blank exercises featuring the word. Wednesday, we had a practice test, a spelling test with the words for the week, but it didn't count. Thursday, we had more exercises. Friday, we had the real test (which sometimes also contained words from previous weeks or even earlier grades). By fourth grade, my teacher dispensed with everything but the final test on Friday. It was up to us to look over the list of words for the week in our textbook. As an excellent speller, I never bothered to look at the list. I was confident I could spell any word the teacher had on the test. But one week, she tripped us up. Instead of giving us the words in the format usually used, she put the sentences on the blackboard in a fill-in-the blank format where the words would go. We had to know what words were on the list. With about 10 words for that week, I was lucky if I got 50%. And I could tell that other kids in the class were struggling, too. I actually remember one word for that test that everyone who didn't study the list had issues with - "o'clock". Most people only think of "o'clock" pertaining to only times that are on the hour, like 11 o'clock. Somewhere before that day, one teacher or another had taught us that "o'clock" could be used for other parts of the hour, but we still didn't think this way. On Friday mornings at 11:30 am, our class and a few others at the school got together in the auditorium for a musical program called Chorus. The spelling test was earlier in the morning. The sentence was, "We will have Chorus at 11:30 _______ this morning." I never even thought about "o'clock". But if she had done the spelling test the usual way, I would have spelled it correctly.

Speaking of spelling, some teachers whose primary purpose in their courses was not to teach spelling would have a rule that "spelling counts" for their test. Off the top of my head, I currently can't remember if the correct spelling is "per cent" or "percent", but on a 7th grade Geography quiz, our teacher announced that anyone who spelled this word wrong would have the answer that included it marked wrong.

Oh man, so many things wrong there.  Units are important, but importance varies.  I make my kids write units for measurements.  You don't want to mix up metric with Imperial.  When the question specifies units and asks in casual language, units are less necessary.  For example, if it asks, "How many dollars does she have left?", units are already implied, and the question is casual.  I know what the units are, the student knows what the units are, and this isn't some professional accounting thing where someone is going to care.  And no, kids don't need to learn how to do it professionally.  Part of the problem with our public school system is that we are trying to teach young children how to function in professions they will never work in.  It's fine to wait until at least high school to be concerned about being perfect with writing units, and a lot of people don't need that until college.  All it will take in college to figure out the importance of units is one or two assignments where your physics professor expresses confusion over which unit your result is in.  (I was good with units before, in part due to public school and in part due to taking a technical writing course in 10th or 11th grade.  Plenty of students forgot units on the first couple of physics home work assignments in college, and then the professor explained how keeping track of units could help sanity check your results.  Magically, the professor never had to talk about units again.)  When I was in public school though, units were a big deal starting around 2nd or 3rd grade.  That was after adding fruits, but it was still absurdly early for most things.

Spelling wasn't bad, but tricks like what you experienced are incredibly inappropriate.  Unless the point of school is still to train 100% obedient factory workers (which was always an immoral, fascist agenda even when it was the point), punishing students for not adhering perfectly to arbitrary rules is just plain wrong and counterproductive.  That's part of why most Americans hate learning so much.  Teachers that put their own personal pleasure and the obedience of students above learning shouldn't be teachers.  Yeah, students do need to adhere to a certain standard of behavior, so that schools can function effectively, but giving students bad grades for writing their answers too soon?  The point of grading is to measure proficiency in the subject being graded.  It is unethical to give students a grade that represents their proficiency as lower (or higher, for that matter) than it actually is, and any teacher who does so intentionally has no business being a teacher.  I don't know how teacher education works, but this tells me ethics needs to be a significant component.

And likewise, making students guess what the words were was just as unethical.  Is the point of spelling to know what words go on the blanks or to be able to spell specific words, when you know what those words are?  Punishing students who are already skilled for not wasting their time on unnecessary work is pretty horrible.  That's part of the reason I was failing 6th grade.  I got punished for using my time more efficiently and for having already learned a lot of the subject matter.  That's not effective education, and teachers who feel that is an appropriate way to teach shouldn't be teachers.

I taught a Video Game Design college course for a few years.  Part of the class was very short (1 page) essays on specific principles of video game design.  My policy for grading was that if your spelling, grammar, punctuation, or whatever was so bad that it was difficult for me to read your essay, I wouldn't accept it, otherwise I would grade it entirely on content.  The fact is, my class wasn't an English class.  It wasn't my business to teach or evaluate your English skills.  And, in a major (Computer Science) where employers actually do look at transcripts and individual course grades, it would be completely out of line for me to represent your English skills in your grade.  If I was looking at hiring someone, I wouldn't want English skills represented in the grades of classes where those skills are not integral to the class.  (I will darn well take into account English errors on your resume though.  If you struggle with English syntax, how can I trust you won't struggle with syntax in programming, where it matters far more?)

This gets into the same thing as teachers giving extra work to teach students a good work ethic.  If you are teaching a math class, you should darn well stick to teaching math.  If you want to teach me a good work ethic, start offering a "work ethic" class.  My third college physics course (basic physics, because the first and second ones didn't transfer) was a bunch of stuff I already knew (from the first one that didn't transfer).  I had a couple of much harder courses, so I slacked on the workbook for the physics class, which was basically just for practice but still graded.  I got a C in the course.  I got an A on every test, and I got the full 100% on most of them.  My grade sheet was mostly 0s, punctuated with 98%s and 100%s for the tests.  I also got an A on the final.  I can tell you, it is a dead giveaway that your grading system is completely broken (or you need to investigate a student for cheating...) when a student can get almost perfect scores on all tests and the final but get a course grade that suggests that student only sort of knows the material.

The point of grades is to reflect proficiency and learning, not to rate a student's dedication, work ethic, or obedience.  If you want rate those, grade them completely separately.  Maybe employers in other fields would care more about college transcripts if those grades actually meant something.

Another thing that tripped me up in school was poor handwriting and, when margins were not marked on the paper, not adhering to margins. I remember doing a booklet in 4th grade (same teacher who put the fill-in-the-blanks spelling test on the board) and adding a lot more than what was expected into it. We were not supposed to get A's unless we did extra work. I fully expected an A on this project. Instead, I got a C because of poor handwriting and disregard of margins.

In 6th grade, my elementary school tried something new. All 4th, 5th, and 6th graders went to a different teacher for each academic subject. With five different teacher personalities to deal with instead of one, I had the potential of five times the trauma. My 4h grade teacher taught social studies to all of us. We actually got along pretty well that year. She was in charge of choosing the students for the prestigious role of working in the school supply store that year, and she chose me because she knew of my conscientiousness.

For English composition and reading, we had a woman who was much loved and usually very kind. But one day, she assigned us essays. She gave us a standard sheet of notebook paper on which to start it. This is the type of paper that has a margin line down the left side, and you could see the margin for the reverse side to keep within the right side margin. She told us that if we needed an extra sheet of paper to ask her for one. I did. She gave me a sheet with no margins drawn on it. In my excitement, I wrote all the way across the page. At the end of class, she came around to collect my papers. When she saw that I had not observed the margins, she tore up my paper (I don't remember if it was both sheets or just the second one) and told me I had to come in after school THAT DAY to finish it. (Incidentally, we were a small neighborhood school where everyone walked to school, so no worries about missing the school bus. Also, this was a morning class, and most of us went home to lunch prepared by our stay-at-home mothers, so I could tell mine then that I would be home late.) I was quite upset because my Girl Scout troop was going to a special program at a library right after school. The teacher would not listen to my argument. What I remember is getting the essay finished quickly and just making it to the car pool departure point for the library trip. A 5th grade girl in my troop also almost missed the trip due to being detained by the same teacher.

I wondered later if giving us the sheet without the margins marked was part of a test to see if kids remembered to keep margins even when the lines weren't marked. As I said, this teacher was usually very kind-hearted, and this type of action was out of character for her. I remember that later in that school year, my grandmother became quite ill, and to take my mind off that, I concentrated especially hard in this teacher's class and ended up with A's from her on my report card.

Here's another time I had my grade knocked down due to neatness, or rather, lack thereof. When I went to computer programming school right after I graduated from high school, I did well. Our last segment was on COBOL programming. Not only did we have to write a program to fulfill the objective of the assignment, we also had to put together a notebook of documentation, similar to the well-researched booklet I got a C on due to poor handwriting and lack of margins in 4th grade. I got a 90 on it. The teacher, who was also the head of the school, told me that he took 5 points off for messiness. I don't remember what I did that was so messy. I probably had a lot of messy typewriter erasures. Or I handwrote everything and didn't observe margins.

Incidentally, I eventually went to college at night from age 22 until I graduated about a month before my 30th birthday. I don't think I ever got a lower grade than I should have gotten due to neatness, poor handwriting, or lack of margins. I did have one professor whose criteria for our term paper was that it had to be two copies submitted on 8 1/2 by 11 paper. I had that size paper at home, but the place where I expected to have the second copy made used 8 by 10 1/2 paper. I explained the situation to him, and he said the smaller size would be acceptable. 8 1/2 by 11 is such a standard size, he probably wrote it out of habit. Maybe his concern was that he didn't want notepad or legal size paper.

In college, many of my professors used rubrics, which were available to the students, for grading.  I had a few who wanted reports with certain formatting requirements, but the rubrics were always very clear about what those were.  Further, they specified exactly how much of the grade each element was worth.  The only way I ever would have gotten a C or lower for poor formatting is if the formatting was so horrible that it was unreadable.  Formatting was typically worth no more than 5% in classes that weren't about writing specifically.  In the two English writing classes I took, formatting was typically worth 10% of the grade, so again, a C never would have happened unless the paper was so unreadable that I lost points elsewhere.  I also took I think three creative writing courses (the teacher rotated through three course titles, so it could be repeated multiple times), and the only time technical skills were ever graded was when they interfered with reading.  (And the teacher of those courses also taught the technical English courses where that stuff did matter, so she easily could have docked points for poor technical skills.)  The only time she ever complained was when a student brought in a full page of single spaced text that was not formatted into paragraphs.  It was for a peer review session, and most of the students (including me) just didn't read it, because a full page without paragraphs is incredibly difficult to read.  (She did compliment my technical skills though.)

What's ironic about a lot of this is that even 50 years ago, typewriters were the most common way of writing professional papers, where formatting actually mattered, and most typewriters either had hardware constrained margins or adjustable margins.  Margins in hand written documents really haven't mattered for well over 50 years.  Further, lined paper with printed margins was invented in the early 1800s and was in common use for professional applications well before 1900.  And, of course, now days everything written in a professional context is written on computers, where margins and other formatting can be set after the fact, before printing or before publishing in a digital format.  My point is, there is not been any context in which being able to write on plain paper without marked lines or margins while maintaining good formatting is a useful skill since the mid-1800s.

On a related note, I was surfing Reddit the other day, and I came across a collection of memes about things they told you in public school that turned out to be totally wrong in college.  Among those was the fact that perfect formatting is rarely very important in college, despite the fact that public school English teachers like to suggest a paper with imperfect margins or other formatting errors would be entirely rejected.  For most of my college, my professors were pretty laid back, focusing more on my skills than whether or not my attention to detail was perfect.  In my English writing classes, they did want a specific format (APA, I think), but the teachers admitted that the only value this had was teaching us to maintain consistent format, and APA was used because it is well documented.  In my CS courses, the only thing professors cared about was consistent formatting, and my Technical Communications professor even showed examples of papers with different formatting from the same journal.  The fact is, very few fields care what formatting you use, so long as it is readable and consistent.  And in fields that do care, journals typically have editors who will convert submissions to the desired format, if they even care that much.

But yeah, a lot of skills they teach or otherwise require students to develop in school are worthless skills that waste time that could be spent on useful education.  (I am not talking about skills like writing cursive, which is actually still useful, though mostly in the context of the fact that a lot of moderately older documents are written in cursive, and learning to write it better cements the ability to read it.)  I think part of this could be solved merely by requiring all public school teachers to develop grading rubrics for all assignments and for each school "term" (year or semester, depending on the school).  Because making formatting be worth 25% of the grade for writing is absurd.  Even college English courses generally don't grade formatting at 25% of the grade.  Of course, this would also help with issues like your spelling test.  If the rubric says 100% of the grade is based on how the words are spelled, that gives students (and their parents) the ability to call out the teacher when the spelling test is based on the ability to recall the words rather than the ability to spell them.  Honestly, to me this is more evidence of how incredibly unscientific and frankly chaotic and unorganized our public education system is.  If a teacher cannot produce a document showing precisely what students are being graded on, the entire grading system is utterly worthless, and if the grading system is worthless, how can anyone expect the system to have or meet any quality standards?  (And yes, I have produced fairly precisely defined rubrics for every course I have taught.  I honestly wouldn't feel comfortable trying to grade students without well defined grade requirements.)

What is a "rubric"?

In an educational context, it is a document outlining how an assignment is going to be graded.  I've seen a few different formats.  One of my professors really liked a grid design, where one side would be labeled with elements.  In papers for the course, elements might be formatting, effectiveness of communication, and understanding of the subject matter.  The other axis would be labeled with quality levels, like "Excellent", "Good", "Acceptable", "Poor", and "Unacceptable".  Each element would have a percentage associated with it, telling how much of the grade it was worth.  Formatting would generally be 5%.  Effectiveness of communication might be 10% or 20%.  Understanding would be the remaining 75% to 85%.  Similarly, the quality would be labeled with values.  The grid boxes would have text explaining what constituted that quality at that element.  So "Excellent" in formatting might say, "Followed all formatting requirements.  Used good spelling and grammar".  "Good" might say, "Followed nearly all formatting requirements.  Used good spelling and grammar."  "Acceptable" might say, "Followed most formatting requirements.  Spelling and grammar don't significantly interfere in readability."  Rubrics for programming assignments would typically just be a list of requirements.  If there were more than two or three requirements, they might also be labeled with percentage values.  Often programming assignments included one or two extra credit requirements that would always have a labeled value, even if the rest didn't.

My rubrics work a little differently.  For my Video Game Design course, I have two sections.  One is "knowledge" and the other is "application", and each is worth 50% of the grade.  "Knowledge" is a list of principles I want students to demonstrate an understanding of.  This is typically done with a short essay (one to two pages), and I allow students to cover two or three related topics in a single essay if they like.  My grading sheet just has check boxes, and when a student demonstrates a proficiency, I check that box.  I have also allowed students to pass off knowledge proficiencies by discussing a topic with me in person or by giving a presentation on it in class.  The application proficiencies are passed off with projects, by demonstrating the correct application of the principle.  The course has two projects, a table top game and a video game.  These are typically sufficient to pass off the application proficiencies, for students that are actually paying attention.  I do allow students to pass them off outside of their projects though, by explaining to me how their projects failed to demonstrate a principle and what they could do to fix that.  (Being one of the best ways to learn, I avoid punishing failure, but I still require students to prove they understand the subject matter.)  The rubric for my assembly language course is just four lettered sections (A, B, C, and D), and to get a particular grade for the course, you have to demonstrate all of the skills listed in that letter section and all lower grade letters.  The exception is the A section, which includes a handful of pretty intense things, so I only require around half of them to be completed to earn an A.  The course also has a final project which is not directly graded but which is mandatory to pass the course.  I encourage students to take advantage of the final project to demonstrate any proficiencies they haven't already gotten checked off.  It works surprisingly well, though in that class I give the students little direction in how to pass off the proficiencies.  At the beginning of the semester, I tell them they can do it however they want.  I point out that some past students wrote a single program for each, while others wrote fewer more complex programs that demonstrated many proficiencies in one shot.  Initially students struggle a little with this freedom, but once they get past that, they do excellent work.  I find students are more motivated when they decide their own assignments, because they take ownership and they care more about quality.  I've had students in this and another class graded similarly come to me weeks after presenting an assignment to show me how they have improved it, even after they have gotten full credit.

Anyhow, long answer to a short question, but I think that should cover it.  There are a lot of ways rubrics can be done, but the important part is that they put the rules the teacher is going to grade by in writing.  Good teachers will generally give students access as well, so that students know exactly what is expected.  When teachers don't make their rubrics available to students, it allows them to choose not to follow the rubric, without any accountability.

Hi rybec!

Sorry I'm late to the party. Nice to see activity here.

One critical element to this discussion has not been raised yet. With respect to the broader topic and the narrower previous topic is whether this 'pedagogical perspective' is even supported by the research. I've read a number of studies on student success and engagement on google scholar that suggest their their approach is antithetic to student achievement. Without citing the sources, the consensus seems to be that futility, unduly harsh punishment for small mistakes, and nonresponiveness to student concerns adversely effects academic outcome. These and other practices permeate the KA curriculum.

Is there anything in the literature that wound even remotely support the proposition that adding a reset button or otherwise making the missing of a question less punitive bad for learning? I'm rather sceptical of their premise.

In response to eltonprice:

Actually, that's a lot of the point of this.  I seriously doubt KA is using an evidence based learning approach.  Typically, when the word "pedagogy" is used to describe a teaching model, underneath the hood is a bunch of arbitrary ideas basically made up by those using them.  (I find that people using more evidence based approaches to education avoid calling their approach a "pedagogy", because they don't want to be associated with an education system where everyone thinks their own system is best and no one bothers to research what actually works best.)

I've read a fair amount of the research, and what you are saying is right on point.  The problem is that only a small portion of the research comes from the education community, and the education community absolutely refuses to consider, for example, psychology research (and they even prefer to blow off research done by other educators, when the results don't agree with their own "opinions").  There are also a whole bunch of small, independent schools and a handful of public schools thinking outside of the box, but educators use the same argument I am using against their pedagogies against those schools.  For example, a few U.S. schools are using linked learning, where first they teach something, say some math technique, and then they apply it in a project, for example designing a bridge.  (Students decide on the projects, the school invites experts to teach the students how to do the stuff, and then the teachers work with the students to apply the learning, and the experts review the results and provide feedback.)  Linked learning is having rather extreme success where it is being used, but we are talking about schools of maybe 200 students max (one highschool, if I recall) and typically more like 40 or 50 (private schools).  Small sample sizes makes it so you cannot generalize the outcomes.  You also have "unschooling" style schools that basically give students resources and let them decide what they want to learn.  In those, teachers teach what students ask them to.  Again, the approach seems to be very successful, but these are small number of private schools with only 10 to 50 students each.  I watched a TED talk on project based schooling, where students do a lot of hands-on work and learn the various subjects in the context of real life projects.  Again, very successful, but there are only a few schools that do it, and they have fewer than 50 students each.  Educators are right to suggest that these techniques don't have enough evidence behind them to generalize them to the entire population, but neither do their own pedagogies.

So, what I want to know from KA is, what is this mysterious "pedagogical perspective" you speak of, and what evidence based science does it come from?  Or, are you behaving just like the U.S. education system and just pulling ideas out of your [donkies] and pretending like that somehow means they are good?  It's ironic how incredibly egotistical American educators are, when we spend more on education than most countries that are performing better than us but barely even perform above average for developed countries that include those currently in circumstances that don't even allow their education systems to operate.

(I did a ton of research on education performance rankings, spending, and salaries a couple months ago.  The U.S. is above average for performance rankings, but those rankings include a lot of countries that are only barely developed and many currently in the middle of political turmoil that has crippled or straight up shut down their schools.  We spend significantly more on education on average, and most countries that perform better than us are spending less.  We also have significantly higher salaries for teachers than the average, and again, most countries that perform better than us pay teachers less.  Also, several countries that pay teachers significantly more perform worse than us.  Notably, Japan is near the top, and they spend around 75% per student of what we spend, and their teachers get paid significantly less.  Of course, Japan has a major cultural advantage, but teachers are in a unique position to influence culture, and somehow U.S. teachers have failed to convince students that learning is important.)

And you know what?  I am fulling willing to admit that they know better than me, if they produce the evidence supporting their decisions.

Here is what the literature actually says: Children have learned historically primarily through play.  In play, the cost of failure is typically very low.  Play also tends to motivate children to try again, often many times, when they fail, especially when they are playing with others whom they may want to impress or compete with.  The most common learning pattern during play is trial and error.  The child wants to accomplish some goal and attempts different solutions until one is successful.  The cost of failure is so low that there is no disincentive to even hundreds of repeated attempts.  The second most common approach, which I call the "walk-through" approach is watching someone else do something and then trying it themselves, often multiple times either to hone technique or to adapt to personal differences (height difference, for example).  In my own courses, I try to use both of these.  In one course, I show students one way of doing something (walk-through), and then I ask them to apply the same thing to a different problem of their choosing (trial and error), and I ask them to pay close attention and report to the class a challenge they ran into.  The result surprised me.  My students played with the problem, looking for challenges to report.  A few even looked up common pitfalls on the internet and then reproduced them to demonstrate them to the class.  (This was awesome, because those pitfalls were the discussion of the next class period, and my students were all pretty familiar with them by then.  My goal was to give them some exposure.  They ended up teaching themselves 90% of what I was going to teach.  This was the class where the lowest grade was an A-, largely because I managed to setup a learning environment where students were highly motivated.  I can't promise I could do this again, because we are talking a sample size of around 14, and maybe I just got lucky with 14 exceptional students, but I basically applied educational psychology and the results were way better than I expected.)

Anyhow, within a teaching environment, the lower the cost of failure, the more motivation students have to keep trying to learn.  That's what the research says, over and over again.

Here's another elementary school horror story that I think could have been better handled by a thinking-outside-of-the-box teacher. Instead, because I had a no-nonsense teacher, I got a very poor grade on the assignment.

We were learning nouns, verbs, and other parts of speech. One day, the teacher handed out a paper containing a story. We had to identify either the nouns or the verbs, I forgot which was which and identified the wrong ones. It was returned to me with every answer marked wrong. What I think an understanding teacher would have done was to notice that one of the brightest kids in his (yes, his, my first male teacher) class had done just the opposite of what she (me, myself, and I) should have done. He should have given me some credit for at least knowing how to distinguish between nouns and verbs. No, I should not have gotten 31 right for a perfect score, but I should not have gotten 31 wrong either. (Actually, I remember getting a D instead of an E [which was our school district's failing mark, not F], so there may have been other parts to this assignment in which I did well enough to raise my grade.) Of course, I never got my nouns and verbs confused after that, but even if an understanding teacher had given me half credit rounded down to 15 right and 16 wrong, instead of 31 wrong, I still would have never forgotten which is which again.

If I was a teacher in that scenario, I would offer the student a second chance.  It is indeed obvious from that, that you can differentiate.  There are two possibilities at that point.  One is, you are confused about which is which, but you know the difference.  In this case, it is my fault for not teaching you more clearly.  If you can differentiate, you were obviously paying enough attention that if I had done my job better, you would have known which was which.  The other possibility is that you accidentally marked them the reverse of what the assignment said, but you do know which is which.  Accidents like that do happen, but I can't give you full points even if you explain the mistake.  So, I would start by asking you about it.  This should be sufficient to determine fault.  If I am at fault, I will go over it again with you, to correct my poor teaching.  If you are at fault, I still can't justify giving you even 50%, because giving you a grade that says your proficiency is lower than it actually is, is no more honest than 100% when you didn't demonstrate proficiency.  So, in either case, I have you redo the assignment, asking you to carefully make sure you answer how you intend, and I grade you on that instead of the first try.

This is why I generally allow my students multiple tries, so long as I have time to grade them.  The fact is, sometimes failure to understand is the teacher's fault, and I try to account for that in my grading policy.  Even when it is not, doing poorly on an assignment does not necessarily mean that a student is not proficient.  It is widely accepted (and then summarily ignored) that tests are a poor evaluation method, because many proficient students will still do poorly on tests (and some like me will reverse engineer the tests and learn how to do things I didn't know how to do until I took the test; yes, I actually did this, when I had to take a college placement test two extra times, because I was not able to afford to start when I thought was I going to, and I did better on each successive try, because I learned from the tests how to do the problems on them).  This is actually true of most evaluation methods.  The best way to alleviate this is actually to reduce stress.  In other words, if you want students to do better on evaluation, allow them multiple tries.  Most students will do better on the first try, when you eliminate the stress of single-try evaluation.

And yeah, the best programmers are the ones that are not afraid of making mistakes.  The fact is, you are going to make mistakes.  The sooner you understand and accept that, the easier things will be.  In my courses, I make a point of not scripting my coding demos/walkthroughs.  I do make mistakes, and I have my students help me find them.  And then, I point out that even experts make mistakes, and typically a lot of mistakes.  The difference is that experts don't let mistakes bother them.  Making and correcting mistakes are just part of the job.  You write some code, you test it, you correct the mistakes.  (In CS, students often get discouraged when they see professors effortlessly write perfect, bug free code every time in class.  These are typically scripted, and the professor has the code in the teaching notes.  In real life, that's not how it works, and I want my students to see that, so they don't get too discouraged by their own inability to avoid making mistakes.)

The most nervous I have ever been when programming was two occasions where I wrote some very complicated code that I could barely fully grasp (and I had over 20 years of experience at the time), and then it compiled and ran without any obvious errors.  The reason I was nervous is that there are three kinds of mistakes.  One causes the code to just not compile.  Most of the time these are easy to diagnose and fix.  The next allows the code to compile but causes the program to obviously misbehave.  These are often harder to diagnose and may or may not be easy to fix.  The last allows code to compile and run correctly 99.9% of the time, and you will never find it in testing, but when you ship it to 10,000 customers, a handful will have serious and possibly expensive failures.  Since you can't fix a bug you don't know exists and can't reproduce once you do know it exists, these are incredibly hard to diagnose and fix.  And since experience will tell you every complex piece of code will initially have some bugs, if it compiles and runs correctly on the very first try, there must be a bug of the third, and worst, type hiding in there where you will never find it until it causes someone serious trouble.  (The first piece was written in a functional programming language, where if you get it to compile, it is more likely to be bug free.  Knowing this does not alleviate the nervousness.  The second was in C, doing complicated memory things that are very prone to bugs, and indeed I did eventually find a bug.  It was so trivial and not at all related to the complicated stuff though, so that also didn't alleviate the nervousness.  (For those familiar, it was a simple buffer overrun due to an off-by-one error in a loop, that had nothing to do with the double pointers and dynamic memory allocation I was using where the bug should have been.)  Later I reviewed the complicated portion of the C code again, and it was indeed bug free, but it shouldn't have been on the first try, because that just isn't how programming works.)

I'd like to first acknowledge the hole-digging analogy by hanz64. I think that was spot on. Also, thank you rybec for your excellent analysis of the counterintuitive, indeed antiscientific, approach to khan academy's 'pedagogical perspective.'

Another point I don't think has been touched on is the lackluster, boilerplate explanations top the individual exercises. They're either overly technical or completely nonresponsive to the challenges of learning the objective. For example, the first three bullets will be identical and the last will be insufficient modified to apply to the current problem. Furthermore, the might be a set of similar problems with different approaches necessary to solve them. I've wasted countless hours oscillating between approaches for lack of understanding why, for example, you would take the negative integral when the bounds are switched. Of course if a and b are in the proper order, you would NOT take the negative. The template explanation will not differentiate.

I propose a community response area for each question similar to the public q and a section sal set up for himself after each video. There, learners can ask specific questions about the example sal's explaining and very the answers they need to achieve their learning objective. And thank God for that, since more often than not, sal's own videos are corrected via screen text.

Finally, I would be remiss if I failed to point out that I've been reviewing material for the weeks that I've already achieved level 6 on. Five of the six ports of the unit test I'm preparing for on calculus bc have already been completed when I did ab. See, I'm terrified top take the test knowing that I'll more than likely hey something wrong and then be roundly punished for making the effort. This I'd material I already know, that Khan academy KNOWS I already know, yet a simple mistake will send me back to the practicing station, a punishment I don't think I deserve, which will frustrate me, which will invariably cause me to make stupid mistakes through my haze of frustration, and on and on this chain reaction goes. On what planet is this a good idea?

I haven't gotten to calc yet (in KA; in real life, I've done calc through Multi-variable Calculus, which was quite intense), so I have not seen this issue, but your solution of a comment section like they have for the videos is a really good idea.  If the examples are valid but just difficult to understand, others will be able to help, and if the examples are invalid, comments would at least allow learners to know that and get better help learning, and they could help KA staff fix issues, if they care to do so.

And yeah, I feel the same about tests now.  I have been sitting on the fractions test for over a week now, afraid to take it for fear of punishment.  I really want to just get 100% on all of the math stuff, but even trivial math (I do fractions in my head without any trouble) is anxiety inducing, because all it takes is one math mistake, one keyboard failure, or one slip on the keyboard...

I would like to point out and expand upon a discussion item that was had on the last post (from three years ago) on this topic. 

Khan Academy is used quite often by schools, especially in these current times of physical isolation and digital classrooms, to fill in gaps of education and prepare students for rigorous tests (such as the AP). As basically everyone has said, many teachers have "Pedagogical" rather than "Evidence-based" (which is a pedagogy in of itself, but that's beside the point) approach to learning. Those teachers might believe that Khan Academy has it exactly right and that students must be punished so they are able to mitigate every mistake they make (even ones due to UI or simple miscalculation as mentioned in the official response). Khan Academy might be bucking off their philosophical values (as others reference Salman Khan might have less involvement with the company now) in order to appeal to those, rather strict, teachers. Whether this represents a majority consensus of US teachers or some marketing calculation that those 'strict' teachers are more likely to utilize Khan Academy than other kinds of teachers, it seems that Khan Academy has decided the best economic condition is to ignore calls for change.

That being said, that doesn't mean change is impossible. Khan Academy has a huge amount of customization for their users and teachers. What their 'pedagogical beliefs' regarding their mistakes is an option, that educators and users can turn off and on, not a required feature. Khan Academy, after all, is intended to be a platform for learning, not an exercise in philosophical or psychological debate. There might be any variety of differing opinions when it comes to how education should be handled, which is why the individual should be provided with options. 

If Khan Academy truly wants to have broad appeal for individuals to learn at their own pace, in the best way for them, then they would be responsive to ideas that differ from theirs. Providing that option would be of little cost to them, I can't see 'strict' teachers or customers leaving the platform because an option was provided that they disagreed with.

I'm going to make another post with a full refutation of their pedagogy, but I just wanted to share my analysis.

This is why myself and others have suggested this should be an option teachers, coaches, and parents can disable.  Personally, I think KA should go for the approaches that have been shown to work best, without pandering to outmoded teachers supporting bad teaching methods that harm students and make people hate learning, but if they feel like they can't get by without doing that, at least they could make it flexible for those of us who are not unfortunate students of such teachers.

HANZ64 I would prefer to not mention the name of the school I attended for seven years (kindergarten through sixth grade) in the late 1950s/early 1960s. I'll just say it was a public school in a large school district that got larger each year due to the maturing of the baby boomers. Our school was a small one with a stagnant and sometimes shrinking population because we were in a neighborhood of rental apartments and small starter homes. Upwardly mobile families moved out.

We had the same principal the whole time, and she had already been there for a few years before I started there. She was still there when my family moved away from there when my youngest sibling was still elementary school age, three years after I moved on to junior high. There was always a high turnover of teachers each year. Some of it was due to women teachers leaving to have babies or because their husbands got jobs in other parts of the country. But many of them were due to transfers to other schools in the district. My 1st grade teacher stayed there one more year after I had her. I always wondered what happened to her. She was fairly young, still in her 20's when I was in her class, but already married. Fifty years later, I decided to Google her. I figured she'd be in her 70's and thus could still be alive. Since she was already married, I didn't have to worry about knowing her married name, unless of course she had gotten remarried or gone back to using her maiden name. Her name was not something very common. To make a long story short, I found her. We spoke on the phone once, which was all I wanted to do. But as soon as I mentioned the principal's name, her tone changed. She didn't quite come out and say it, but I think she requested a transfer to another school to get away from that principal. She spoke highly of the principal at the school she transferred to.

My children use Prodigy, and they do really enjoy it.  It's a fairly decent game, and the learning elements are not too bad either.  It's not ideal though.

I have a background in video game design, and I have taught college level video game design for several years.  I started teaching as a student, for my internship.  (The CS department at my university was pretty flexible for me, because I started with 20 years of experience.)  One of the conditions the professor acting as my mentor put on me for teaching this course as my internship was that I include some emphasis on educational video games in my curriculum.  In studying both video game design and educational game design, in the process of developing my curriculum, I discovered a few critical things about educational game design.  (I already had a background in both video game design and gamification, but I wanted to do it right, so I reviewed and did deeper study anyway.)

The first discovery, which I honestly already knew, is that the vast majority of educational games are terrible.  The first generation of educational games rewarded students for doing test-like work with very short periods of game-play.  The actual work was just as tedious as class work or homework, and the game-play portion was like a shorter recess that you only get if you perform well.  The work was a gate to the play, and the mental impact of this was that it taught students that school work was something bad that you had to deal with to gain access to something desirable.  Basically, first gen educational games were school work interrupted by brief periods of game-play, with the effect of teaching students that learning is a chore.  The second generation of educational games went in the opposite direction.  They were primarily games, periodically interrupted with educational work.  Where kids hated the first gen educational games, they loved the second gen educational games.  The second gen games were bad at teaching though, and they still taught students that learning is an undesirable chore that you just have to do.  The educational work was still used as a gate.  To access the next part of the level, you have to solve some math problems.  Prodigy is a second gen educational game, in this context.  The school work interrupts game play, and it is a gate that blocks progression.  Prodigy is better than the early second gen educational games, because it actually teaches.  Early educational games only presented the player with problems and sometimes solutions, but it didn't teach how to get the solutions.  Prodigy has hints and such that teach, by explaining.  Fundamentally though, it is still a second gen educational game.  First gen educational games were actually just tests with game elements, and second gen educational games were/are just video games with educational elements.  Neither is good for education.  In both, learning is presented as a chore or even a punishment.  Students hate first gen ones, because they are mostly work and very little fun.  Students love the second gen ones, because they are mostly fun with very little work.  Neither integrates the educational element into the game-play.  The education interrupts game-play, and the motivation to learn is merely to get past the educational elements so you can get back to the game.  They present education as a chore you have to do, if you want to continue to access fun.

The second discovery I made was that games that are not intended to be educational actually tend to be better educational games that ones that are intended to be educational.  The poster child for this is Minecraft, but it wasn't the first, by a long shot.  It has been the most viral though.  Minecraft doesn't teach math.  Minecraft doesn't teach electronic logic.  Instead, it provides such strong motivation to learn that players put their own effort into learning these.  Minecraft also facilitates the learning of principles of interior decoration, very basic architecture, landscaping, and all sorts of other things, and again, players go out of their way to learn these things, because the games motivates them to want to learn.  Minecraft is a far more successful educational game than any game created specifically to be educational, because it provides motivation to learn.  And the quality of learning is much higher, because when students want to learn, they learn more and retain more.  Like I said, Minecraft isn't the first, and there are plenty more.  There are MMOs with an emphasis on economic concerns that provide motivation to learn business principles, economics, some fairly advanced math, and accounting.  Even World of Warcraft provides opportunities and some motivation to learn economics.  (Many years ago, when I actually played, I took control of the copper market on the server I played on, establishing a monopoly and driving the price up to about 4 times its initial average.  My success required an understanding of supply and demand economics, and it also turned out to be a really interesting social experiment.  Despite the high availability of copper, I was able to drive prices up, because there weren't enough players willing to spend the time mining it to drive prices back down.)  The solution to the problem of educational games is to make games that help students want to learn, rather than making learning a chore.  This is a lot harder though.

Anyhow, the educational game industry doesn't understand even the most basic principles of video game design, and that's why educational games either are hated by students or educate poorly and always make students see learning as something undesirable but necessary.  The games that are successful at educating are the games where the educational elements apply directly to the game-play, such that students want to learn, to improve their quality of play and their success in the game.  In Minecraft, math helps you mine, craft, and build more efficiently.  You want to calculate how many blocks your building needs, so you know how many to gather.  You want to know how much of an ingredient you need to craft the number of an item you want to craft, so you can gather it in one shot.  You know that you are going to have to smelt a bunch of cobblestone to build your vault out of stone brick, and you know each coal will smelt 8 blocks of cobble.  You don't want to keep going back and fourth to your mine to get more coal, so you calculate how many bricks your vault needs, you divide that by 8, and then you go down, mine that much coal, and come back.  (Or, you make a tree farm, to make charcoal, and you do a little more math to figure out how much fuel you are going to need to make enough charcoal.)  The game doesn't try to force math down your throat.  It gives you a strong reason to learn.  Imagine a Minecraft mod that adds a math tutorial.  You start smelting cobbles, and it offers to teach you how to calculate your fuel requirements.  Or one that offers to teach you electronic logic and how to build redstone logic gates, when you craft a redstone component.  Kids are already going to Google to learn this stuff.  All Minecraft needs to be the best educational game ever created is a little bit of teaching, so students can learn within the environment of the game, instead of having to Google everything.

Anyhow, Khan Academy isn't trying to be an educational game.  It is trying to gamify education though.  Early Khan Academy did a reasonably decent job of this.  Yes, it wasn't perfect, but it worked fairly well.  Modern Khan Academy fails at this though.  The intent of gamification is to influence the behavior of people to get them to behave in ways they wouldn't naturally behave.  Gamification doesn't mean making something into a game.  It means adding game-like elements to create the same kind of motivations that motivate people to desire to perform well in games.  Khan Academy gamifies to motivate people to want to learn, and it actually does this better than first generation educational games.  But, it is starting to fail at this, because gamification can easily be defeated by having just a few elements that artificially get in the way.  In a business that uses a leaderboard to track sales and motivate sales people to work harder, just adding one piece of paperwork they have to do to get their numbers on the leaderboard can substantially reduce motivation.  They know the sales themselves will require paperwork, because it is necessary to record orders, track inventory, and allocate commissions.  If they also have to do additional paperwork to get their sales counted on the leaderboard, which they know could easily be handled as part of processing the sales, there goes most of the motivation for most of the people.  And that's a small thing.  Khan Academy's lack of any way to reset tests (when users know this is merely a design decision, not a limit of the technology), and the huge punishment that lack creates, is way bigger than having to fill in a few blanks every week or month.  In short, this one failure effectively negates nearly all of the gamification motivation Khan Academy has worked so hard to include.  That's all it takes.  That's why educational games are not very effective.  One arbitrary punishment or inconvenience is worth many strongly motivational elements.  That's why we haven't gotten past second gen educational games.  If you don't do it right, you need to have at least 80% game-play to no more than 20% education, just to keep students playing.  Because if the education isn't part of the fun, you need many times more fun to make up for the education, otherwise no one will want to play the game.  (And honestly, this is probably the one element of educational game design the industry actually understands.  Second gen is certainly an improvement on first gen, but it isn't the end of the road.  We can do better.)

Hi everyone,

We discussed and considered a reset button for exercises and unit tests at length, but ultimately decided against using it on Khan Academy. We serve a wonderfully diverse group of users, and we understand that this approach won't work for everyone.

As a reminder, our goal in the Support Community is to foster a safe place for everyone to learn and help each other, and that name-calling and insults harm the Community for everyone. It creates a hostile environment where learners may not feel safe to ask questions or participate in discussions.

We've deleted several comments that were off-topic or insulting to other users or Khan Academy staff. Again, we encourage your thoughtful discussions about how you think Khan Academy can be improved, but please keep your discussions respectful and kind to everyone, in keeping with our Community Guidelines.

@...,

Please keep your comments respectful as Evan Lewis suggested. The staff at Khan Academy desire to keep the atmosphere a wholesome educational environment. Insults and disrespect do not reflect that. A lot of hard work is currently being put into Khan Academy and the staff has their hands full, so they may not reach the conclusion you want when you want it.
Thank you for consideration.

Best,
The Anonymous

As we mentioned earlier, we welcome feedback and criticism, but we also have to ensure that our Support Community is a safe, welcoming place for everyone. Because we're seeing off-topic comments and other responses in violation of our Community Guidelines, we're closing this thread for future comments.